Teaching Computational Geometry

15 directed towards more application-oriented students. Of course the diier-ent possibilities discussed are not the only ones. Probably each person who teaches a course in computational geometry will choose a diierent set of topics and will treat them in a diierent way. Also computational geometry is still evolving very quickly. New algorithms, approaches, problems and applications are discovered all the time. At the moment for example there is a strong tendency towards randomized algorithms for computational geometry (see e.g. the textbook of Mulmuley93]) in contrast to the more traditional deterministic approaches. This will all have an impact on future courses. Again, I would like to point out the importance of computational geometry in application areas like graphics, robotics, and geographic information systems. First of all, techniques and algorithms developed in computational geometry will be useful for people working in these areas. Secondly, computational geometry will give them a better insight in the theory underlying the diierent geometric questions arising. It will save people from reinventing the wheel or from working too long on impossible missions. Computational geometry is deenitely not only a eld of research with beautiful combinatorial and algorithmic results but also a eld with lots of practical impact. Possibly due to the lack of good textbooks it is unfortunately too unknown among practitioners. I hope this will change in the near future.